Cover Image

Not for Sale



View/Hide Left Panel
Click for next page ( 94


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 93
Glossary We have developed a standard glossary of terms and their definitions, the use of which will make comparisons among surveys easier if adopted by the transportation planning profession. The purpose of a glossary is to promote understanding and establish common terminology within the subject area. Some of the terms used in travel surveys present no problem in understanding or interpreta- tion among users (e.g., terms such as "gender" and "age"). Others require definition within a particular study because while there is generally no difficulty in understanding the general mean- ing, the specific meaning must be defined to ensure a common use of the term. Examples of these terms are words such as "trip," "employment status," or "occupation." These types of terms are appropriately defined in each study although there is definitely merit in establishing uniformity in these definitions among studies. However, there is a third group of terms which are in com- mon use but whose meaning may vary among persons in the field. Examples of such terms are expressions such as "pretest," "activity," and "sojourn." It is this type of term that is most appro- priately described in a glossary. In the section below, definitions of terms commonly used in per- sonal travel surveys are suggested. It is recommended that the terms described below be accepted as a glossary for personal travel surveys. The terms are presented in alphabetical order with the source of the definition, where appropriate, in parentheses after each term. To be effective, a glossary must be updated from time to time in order to amend any errors and add new terms. Activity: The main business or undertaking engaged in by an individual, alone or with others (Zimowski et al., 1997a). Note, travel qualifies as an activity. Also, an individual can conceiv- ably be engaged in more than one activity at a time (e.g., eating while watching television). Advance Letter (or Announcement Letter): A letter sent to a sample member (household or per- son) via mail in advance of an attempt to contact the member (Zimowski et al., 1997a). It can be used to provide contact telephone numbers and a website address for obtaining more infor- mation about the survey (NHTS, 2001a). Calibration Experiment: When old and new versions of a survey instrument are administered to different portions of the sample to assess the impact of changes in the questions on responses (Zimowski et al., 1997b). Call Back (Disposition Code): This is a disposition code in a call-history file that shows that the household has requested to be called back. Contact was, therefore, made with the household, but a complete recruitment or retrieval interview was not obtained. Call-History File: A call-history file is the file that houses disposition codes (labels) for each call made, on each call attempt, during the recruitment phase of the survey process. This file lists all households contacted, the disposition code showing whether the household was recruited 93

OCR for page 93
94 Standardized Procedures for Personal Travel Surveys successfully, and the number of call attempts required for this to eventuate and shows the number of households that declined to participate in the survey. It also contains other infor- mation, such as the type of recruitment--e.g., whether a cold call is made or the intercept recruitment method is adopted--and records the time, day, and date when each call was made and, importantly, the telephone number. It may also include some demographic information about the recruited household (see also Recruitment Interview). Calling Protocol: A procedure followed while contacting sample members. Most calling proto- cols take into account the outcomes and times of previous calls in an attempt to improve the chances of reaching sample members (adapted from Zimowski et al., 1997a). Cluster Sampling: In cluster sampling the total population is divided into clusters of sampling units, usually on a geographic basis, and clusters sampled randomly with the units within the selected clusters either being selected in total or else selected randomly, usually at a very high rate (Richardson et al., 1995, p. 92). Codebook: The collection of all coding frames for a survey is referred to as a codebook (adapted from Richardson et al., 1995, p. 269). Coding: Coding is the translation of respondent answers into machine-readable data. Coding Frame: A coding frame describes the set of codes to which answers to codes are allocated (adapted from Richardson et al., 1995, p. 269). Cognitive Interview: A technique for developing survey questionnaires that focuses on the thought processes respondents go through as they arrive at answers to survey questions. In a cognitive interview, respondents are asked to think out loud as they answer draft survey ques- tions. They may also be asked to respond to a number of follow-up probes to reveal how they arrived at their answers and whether the content or wording of the items should be improved (Zimowski et al., 1997a). Cognitive Laboratory: An experimental setting for piloting the framing, order, and wording of survey items. Cohort: A group of individuals within a population who have experienced the same life event during some specified period in time. A cohort is usually defined by year or period of birth, but it may also be used to refer to the timing of any number of other life events such as year of retirement or year of marriage (Zimowski et al., 1997b). Cold-Deck Imputation: A statistical procedure that replaces missing or incorrect data items in one data set with values taken from a similar case in a different data set. Computer-Assisted Personal Interviewing (CAPI): Face-to-face interviewing performed with the assistance of a computer. In a CAPI interview, the interviewer reads the questionnaire items from a computer screen and records the respondent's answers by entering them into the computer (Zimowski et al., 1997a). Computer-Assisted Self-Administered Interviewing (CASI) or Computer-Assisted Self- Administered Questionnaire (CASAQ): A computer-based instrument that is completed by a respondent without the assistance of an interviewer. This may involve the loan of a com- puter, the installation of the instrument on a computer chosen by the respondent, the use of a public computer terminal (e.g., in an airport waiting area), or access via the Internet. The functions possible are similar to those of CAPI, but greater attention may be given to inter- active graphics, etc., to make the survey easy to use and enjoyable. Computer-Assisted Telephone Interviewing (CATI): Telephone interviewing performed with the assistance of a computer. CATI systems are similar to CAPI systems in that the question-

OCR for page 93
Glossary 95 naire items are displayed online and the interviewer enters the respondent's answers with the keyboard or mouse (Zimowski et al., 1997a). Consolidated Metropolitan Statistical Area (CMSA): A metropolitan complex of 1 million or more population, containing two or more component parts designated as Primary Metro- politan Statistical Areas (PMSAs) (NPTS, 1990). Converting: Recontacting initial refusals one or more times in an attempt to persuade them to participate in the survey (Zimowski et al., 1997a). CPS: Current population survey (Zimowski et al., 1997b). CPS Supplements: Supplements have the same data as the Basic Monthly Files plus supplemental information on topics such as schooling, fertility, immigration, income, and technology use (NBER, 2002). Critical (Key) Items: A set of questions that must be completed by the sample member to classify the case as a respondent. The set typically includes all items that are essential to accomplishing the major goals of the survey (Zimowski et al., 1997a). Cross-Sectional Survey: The collection of data at one point in time from a fraction of the total population (Richardson et al., 1995, p. 34). Disposition Code: A code assigned to each case (member) in the sample that records the most recent or final outcome of the data collection effort (e.g., ring-no-answer, non-working num- ber, respondent refusal). Disposition codes are used to track the status of each sample mem- ber, to monitor and manage the field effort, and to compute response rates (Zimowski et al., 1997a). Disproportionate Sampling or Stratified Sampling with Variable Sampling Fraction: A sam- pling method in which the population is divided into strata, and different sampling rates are used for each stratum. This method of sampling, while random within strata, does not pro- duce a total sample that is representative of the population without weighting. It is also a method of sampling that produces the optimal sample for given cost and accuracy. Driver: A person who operates a motorized vehicle. If more than one person drives on a single trip, the person who drives most of the distance on the trip is classified as the principal driver (NHTS, 2001b). Eligible Units (Eligibles): Sample units who are eligible to participate in the survey because they are part of the target population (Zimowski et al., 1997a). Expansion Factor: The inverse of the achieved sampling rate. It is the factor required to multiply each observation by to expand the sample to the total population. Field Period: The time period during which survey data are collected from the respondents (Zimowski et al., 1997b). Hard Refusal: A sample member who has objected strongly to taking part in the survey (outright refuser). Hot-Deck Imputation: A statistical procedure that replaces missing or incorrect data items with values from a similar case in the same data set. Household: All persons who occupy the same housing unit and share responsibilities and resources. A household may consist of a family, one person living alone, two or more families living together, or any other group of related or unrelated persons who make joint decisions and share resources (adapted from Zimowski et al., 1997a).

OCR for page 93
96 Standardized Procedures for Personal Travel Surveys Household Income: The total household income before taxes in the past 12 months. Includes income from sources such as wages and salaries, income from a business or a farm, Social Security, pensions, dividends, interest, rent, and any other income received (NHTS, 2001b). Household Members: Includes people who live in the household but excludes anyone who usually lives somewhere else or is just visiting, such as a college student away at school (NHTS, 2001c). Household Vehicle: A motorized vehicle that is owned, leased, rented or company-owned and avail- able to be used regularly by household members during the travel period. Includes vehicles used solely for business purposes or business-owned vehicles if kept at home and used for the home to work trip (e.g., taxicabs, police cars) that may be owned by, or assigned to, household mem- bers for their regular use. Includes all vehicles that were owned or available for use by members of the household during the travel period even though a vehicle may have been sold before the interview. Excludes vehicles that were not working and not expected to be working within 60 days and vehicles that were purchased or received after the designated travel day (NPTS, 1990). Housing Unit: A house, apartment, mobile home, group of rooms, or single room that is occupied (or if vacant, is intended for occupancy) as separate living quarters. To qualify as a housing unit, the occupants must live and eat separately from other persons in the building and have direct access to their unit from the outside of the building or through a common hall (Zimowski et al., 1997a). Imputation: A statistical or logical technique by which missing or incorrect data are estimated from other information provided by the respondent and/or other respondents. Imputed val- ues are probabilistic estimates and therefore there is no expectation that individual imputed values are necessarily correct (Zimowski et al., 1997a). Incentive: A gift or payment offered to sample members in an effort to gain their cooperation (Zimowski et al., 1997b). Ineligible Units (Ineligibles): Sample units who are not qualified to participate in a survey because they do not belong to the target population (Zimowski et al., 1997a). Inference (Logical Imputation/Editing): Where values of missing or incorrect data items are log- ically deduced from information furnished by the respondent and/or other respondents. In inference, values are deduced with a high degree of certainty. Item Non-Response: The failure to obtain a specific piece of data from a responding member of the sample (Zimowski et al., 1997a). Journey: A journey is a tour starting and ending at home (adapted from Axhausen, 2003). Home may include a temporary home such as a hotel room, provided the individual slept at the loca- tion for at least one night. Locating Letter: A letter sent by mail to the respondents in advance of the next data collection period in an effort to obtain updated addresses and telephone numbers before the next round of data collection (Zimowski et al., 1997b). Longitudinal Data: Data collected on multiple variables at intervals over time. Longitudinal Panel Designs: Designs that collect information on the same set of variables from the same sample members at two or more points in time (Zimowski et al., 1997b). Longitudinal Weights: Weights designed to be used in longitudinal analyses of data from a panel survey (Zimowski et al., 1997b). Means of Transportation (Mode): A mode of travel used for going from one place (origin) to another (destination). Includes private and public modes, as well as walking.

OCR for page 93
Glossary 97 Missing Data: See Item Non-Response (Zimowski et al., 1997a). Module: An independent part of a questionnaire that covers a single subject or topic of interest or one instrument in a multi-instrument design (adapted from Zimowski et al., 1997b). Multistage Sampling: A sampling technique in which the sample is selected in two or more successive stages (Richardson et al., 1995, p. 89). Non-Response: Failure of the survey to obtain the desired information from eligible sample members (Zimowski et al., 1997b). Non-Response Bias: Non-response bias is a function of the non-response rate and the difference between respondents and non-respondents on the statistic of interest (Keeter et al., 2000). Non-Response Weighting: Post-sampling statistical adjustment to compensate partially for pos- sible non-response error. Statistical weighting to compensate for non-response is different from the post-sampling weighting that is routinely performed to adjust for unequal probabilities of selection (Zimowski et al., 1997b). Non-Working Number: A telephone number that has not been assigned to a unit by the telephone company (Zimowski et al., 1997a). Occupancy: The number of persons, including driver and passenger(s) in a vehicle (NPTS, 1990). One-Time Cross-Sectional Designs: Designs in which sample members are asked to complete a survey once (Zimowski et al., 1997b). Open-Ended Question: A survey question that asks the respondent to phrase the answer in his or her own words (Zimowski et al., 1997a). Panel Attrition: Failure of first-wave respondents to complete the survey in subsequent rounds of a panel survey (Zimowski et al., 1997b). Paper-and-Pencil Interviewing (PAPI): An interview in which the interviewer reads the ques- tions from a printed questionnaire and records the answers directly onto the questionnaire using a pen or pencil (Zimowski et al., 1997a). Person Miles (Kilometers) of Travel (PMT or PKT): A measure of person travel. When a person travels 1 mile (kilometer), 1 person mile (kilometer) of travel results. Where two or more per- sons travel together in the same vehicle, each person makes the same number of person miles (kilometers) as the vehicle miles (kilometers) (NPTS, 1990). Person Trip: A person-trip is a trip by one or more persons in any mode of transportation. Each person is considered as making one person-trip (NPTS, 1990). Pilot Survey: A pilot survey is a test of all aspects of the survey process (Richardson et al., 1995, p. 214). Population: The universe of elements from which a sample is to be drawn. The population is defined in terms of the nature of the elements (e.g., people, households, vehicles); time; loca- tion; and any other appropriate descriptor (e.g., all persons living within the metropolitan region of Atlanta in April 1993). Post-Stratification: The process of weighting stratified data to compensate for different sam- pling rates in the strata. Pretest: A relatively small number of practice interviews used to test and refine individual com- ponents of the survey material and/or field procedures before the main survey is executed (Zimowski et al., 1997a).

OCR for page 93
98 Standardized Procedures for Personal Travel Surveys Primary Sampling Unit (PSU): Units included in the sampling frame of a first level of a multi- stage sampling procedure, such as all PMSAs in the United States. Proportionate Sampling or Stratified Sampling with Uniform Sampling Fraction: Sampling from strata, using the same sampling rate in each stratum. Thus, each stratum will occur in the sample at the same proportionate rate as in the population. Proxy: A proxy report results when someone other than the selected sample member answers survey questions on behalf of that member (Zimowski et al., 1997a). Quota Sampling: Sampling in which a specified number of responses are required from selected groups within the population (Richardson et al., 1995, p. 95). Random Digit Dialing (RDD): Techniques that form samples by adding random digits to the telephone prefixes that fall within the sampling area so as to include both listed (published) and unlisted numbers in the sample (Zimowski et al., 1997b). Random Sampling: A process in which each unit in the population is sampled independently so that each unit in the population has an equal probability of being selected (Richardson et al., 1995, p. 81). Random Seed: A randomly selected starting point for a systematic sample. Recall Attempts (Follow Up): Re-approach of a sample member after previous contact attempts have been unsuccessful in obtaining a complete response or a response (Zimowski et al., 1997a). Recall attempts include contact attempts by telephone, personal visit, mail, or possibly email. Recruitment Interview: The interviewer seeks to obtain consent and cooperation for the survey during the recruitment interview. Household demographic information may be obtained dur- ing this interview, regardless of whether consent and cooperation for the survey are obtained. Refusals: Sample members who refuse to participate in the screening or main data-collection phase of a survey (Zimowski et al., 1997a). There are three main categories of refusals: hard refusals, soft refusals, and terminations. Reminder Call: Recontact of a recruited sample member to remind them of an upcoming event or their required response to a past event. Repeated Cross-Sectional Design: Designs that collect measurements on a population over time by repeating the same survey on two or more occasions. During each time period, a separate and independent but comparable sample of units is drawn from the population (Zimowski et al., 1997b). Reporting Period: A reporting period is the time interval for which a respondent is asked to describe the events or episodes which are within the scope of the survey (Axhausen, 2003). Respondent Rules (Respondent Selection): Rules that are used by the interviewers to choose a respondent from all eligible members within a sampling unit (Zimowski et al., 1997a). Response Rates: The ratio of the number of completed interviews divided by the number of eligible units in the sample (CASRO, 1982). Retention Rates: The proportion of respondents from the first wave that complete later waves of data collection in a panel survey (Zimowski et al., 1997b). Retrieval Interview: This is an interview whereby household travel responses are obtained over the telephone (CATI retrieval of data). Rotate-Outs: Respondents who are dropped from a panel survey at a pre-specified stage in the survey.

OCR for page 93
Glossary 99 Rotating Panel Designs (Revolving): A panel design that collects measurements on a sample for some specified number of periods (waves) after which some or all of the sample is dropped from the survey and replaced with a new but comparable sample of units drawn from the current population (Zimowski et al., 1997b). Rotation Group: A sample of units drawn from the population at the same time and following the same schedule of data collection in a rotating panel design (Zimowski et al., 1997b). Sampling Frame: A base list or reference that properly identifies every sampling unit in the survey population (Richardson et al., 1995, p. 77). Sampling Unit: An element in a sampling frame (Zimowski et al., 1997b). Screening Interview: A preliminary interview used to determine the eligibility of sample mem- bers. Sample members who meet the eligibility requirements are often recruited for more detailed data collection during the interview (Zimowski et al., 1997a). Self-Administered Questionnaire (SAQ): A questionnaire that is completed by the sample mem- ber without the assistance of an interviewer (Zimowski et al., 1997b). With this type of survey, respondents are required to perform the three tasks on their own--to read and understand the questions, to mentally formulate an answer, and to transcribe the answer onto the questionnaire form. Self-completion questionnaires can be distributed and collected in the following ways: Mail out/mail back surveys; E-mailed to respondents and e-mailed back by respondents (or URL address provided to respondents--automatic data retrieval); Delivered to respondents/mailed back; Mailed to respondent/collected from respondent; and Delivered to respondent/collected from respondent. Soft Refusal: This is a respondent who has not objected strongly to participating in the survey. Refusal conversions (Conversions q.v.) are only conducted on those who provide soft refusals to participation. Sojourn: A stop on a journey. Split Panel: In a split panel, the split is between a cross-sectional survey and a panel (Kish, 1985; Raimond and Hensher, 1997). At each wave of measurement of the panel, which may be either a subsample panel or a refreshed panel, a separate, non-overlapping cross-sectional sample is also drawn and surveyed. Stratification: Process in which units with similar characteristics are divided into groups called "strata" before the sampling process begins. Each unit is assigned to one and only one stratum based on prior knowledge about the unit (Zimowski et al., 1997a). Stratified Random Sampling: Random sampling within individual strata. Strata may be sampled at the same sampling rate (proportionate sampling or sampling with a uniform rate) or at dif- ferent sampling rates (disproportionate sampling or sampling with a variable sampling rate). Stratum: A sub-grouping of the population according to some criterion--e.g., households by household income. Survey Period: The survey period is the time interval over which information is collected from respondents. It is the period covered by all the individual reporting periods in the survey (adapted from Axhausen, 2003). Systematic Sampling: Sampling by taking every nth unit from the sampling frame list, usually starting from a random seed.

OCR for page 93
100 Standardized Procedures for Personal Travel Surveys Target Population: The complete group about which one would like to collect information (Richardson et al., 1995, p. 75). Terminations/Break Offs: Sample members who begin but terminate an interview before they provide enough information to be classified as respondents (Zimowski et al., 1997a). Time-in-Sample Effects: Effects of prior reporting on reporting in subsequent waves of data col- lection. These include conditioning and fatigue (adapted from Zimowski et al., 1997b). Time Series Design: A design that collects a series of repeated measurements on one, or a limited number, of variables over a relatively large number of points in time (adapted from Zimowski et al., 1997b). Tour: A tour is a sequence of trips starting and ending at the same location (Axhausen, 2003). Trip: A linked trip is the movement from one location to another for a single purpose by one or more modes of transportation. An unlinked trip (also referred to as a "stage" or "leg") is that portion of a linked trip conducted on a single mode of transportation. When a trip is not qual- ified as being either linked or unlinked, it is usually assumed to be a linked trip. Trip Chain: A sequence of trips starting and ending at home or work in which one or more inter- mediate stops are made. The stops may be made to change mode or to serve an intermediate trip purpose. Trip Purpose: The reason the trip was made. If there is more than one reason for the trip and the reasons do not involve different destinations, then the reasons must be ranked to identify the main purpose of the trip if a single trip purpose is required. Unit Non-Response: The failure to obtain useful information on critical data items from an eli- gible sample member resulting in exclusion of this sample member from the sample (adapted from Zimowski et al., 1997a). Unit non-response may result from eligible sample members that refuse to participate in the survey and also from eligible sample members who could not be contacted by telephone, mail, email, or personal visit. Validation Interview: Interview of a selection of responding households to check on the quality and completeness of the data provided in the previous survey process (adapted from Richardson et al., 1995, p. 247). Variable Fraction Stratified Random Sampling: Random sampling at different sampling rates in each stratum (see Disproportionate Sampling). Vehicle Occupancy: The number of persons, including driver and passenger(s), in a vehicle; also includes persons who did not complete a whole trip (NPTS, 1990). Vehicle Trip: A trip by a single vehicle regardless of the number of persons in the vehicle (NPTS, 1990). Wave: A distinct occasion when data are collected in a panel survey. Also referred to as a "round" of data collection (Zimowski et al., 1997b). Weight: The factor by which an observation or a group of observations must be multiplied to correct their representation in the sample for biases in the sample, usually as a result of non- response and other factors. Weighted Response: A response rate that is calculated using the inverse of the selection proba- bilities as the weights. If different selection probabilities exist within different strata or differ- ent stages of complex sampling designs, the weighted response rate is a weighted aggregation of the response rates in the strata and stages (adapted from Zimowski et al., 1997b).